I sent 200 applications and got nothing — is it the market or my dossier?

What your series says once you count first human replies

A first human reply means an invitation, a call back, a question, or a rejection written for you personally. An automatic acknowledgement does not count. Intervals by the Wilson method, our own calculation.

Why 200 applications are almost never 200 comparable applications

We split our open stock by occupational field and region and counted what appeared in each slice over the last thirty days. Across 101 slices, 200 adverts equal 9.1 months of everything the median slice publishes. In 63.4 per cent of slices, 200 applications exceed half a year of the entire local output; in 42.6 per cent they exceed a full year.

For a reader applying in English the arithmetic is harsher, and this is the number that explains most expat series. If you restrict yourself to adverts written in English that do not demand German, French or Italian at working level, the median slice needs 33.3 months to publish 200 adverts, and 79.3 per cent of those slices need more than a year. Engineering in the Zurich area is the largest such slice in the country and still needs 2.7 months.

The consequence is uncomfortable and useful at the same time. Anyone who has written 200 applications has necessarily left their slice — across occupations, across regions, across seniority levels. That is arithmetic rather than a criticism: there is no other way to reach 200 adverts within a few months. But it means your series is not one test of your dossier. It is a mixture of attempts with very different odds, and every step outward lowers the fit before anyone reads a word.

The market did not shrink — the matching got slower

The official check is unambiguous. In its July 2026 edition SECO reports 13.4 per cent more vacancies notified to the regional employment centres than in the same month a year earlier, and 22.7 per cent more among the occupations covered by the vacancy reporting obligation. Against the previous month there was a fall of 4.4 per cent, which is the ordinary summer movement rather than the state of the market.

What did shift is duration. Among registered unemployed people, 45.6 per cent have been searching for longer than six months, against 41.8 per cent a year earlier, and the share searching for more than a year rose in the same comparison. More adverts and longer searches at the same time is only possible when profiles and adverts match each other worse — not when there are fewer adverts.

For your question that settles one half of it. «The market is terrible» does not survive the vacancy figures, and «I am doing everything wrong» does not follow either. The third explanation is the statistically likeliest one and the only one you can act on: a large part of your 200 went to positions that never fitted your profile in scope or in stated conditions.

When 200 starts to prove something

Twenty applications without an answer prove nothing. Two hundred is the order of magnitude at which the number begins to speak. If your true rate of first human replies were five per cent, a run of 200 with no reply at all would be effectively impossible. At three per cent it happens in 0.23 out of a hundred series, at two per cent in 1.76, and only at one per cent does it become ordinary again, in 13.4 out of a hundred.

The shortest form of that calculation is the rule of three: with no event in 200 attempts you can say with 95 per cent confidence that the true rate is below 1.5 per cent. That is what your series actually establishes. It does not say your CV is bad. It says that the combination of the adverts you chose and the documents you sent sits below any plausible average — and that the cause is in one of those two halves.

If your series is not at zero, the table above is worth more than a gut diagnosis. Six replies out of 200 is three per cent, with an interval from 1.4 to 6.4 per cent. Reading «my CV is the problem» into that is reading too much into it. How many applications it takes to prove that a revision helped at all is worked out on our question «How do I measure if my CV improved?».

The gate that closes before anyone reads your dossier

Part of every large series is decided before the file is opened. Of open adverts excluding apprenticeships and internships, 66.5 per cent state a language requirement, and where one is stated, 21.0 per cent demand two languages at working level. If you hold one of them, roughly a fifth of your own slice is formally closed to you regardless of how good the documents are.

The shape of that gate changes completely by region, which is exactly what people applying from abroad tend to underestimate. In the Zurich area, of adverts with a stated requirement, 56.2 per cent ask for German at working level and precisely the same share asks for English. In French-speaking Switzerland it is 56.6 per cent French alongside 48.7 per cent English; in Ticino 73.9 per cent Italian. English is no longer a bonus in the German-speaking market — it is a condition in about half the cases, which is good news you can only use if your CV states your levels as A1 to C2 rather than as «fluent».

Two smaller items belong in the same calculation. Apprenticeships, internships and trainee positions make up 5.9 per cent of open adverts, and a broadly scattered series collects them without any of them being winnable for an experienced applicant. And the permit line: a Swiss dossier states nationality and permit status, which a US résumé never does. Leave it out with a foreign degree and the person sorting the pile has to guess — in a stack of applications, guessing rarely goes your way.

The same file sent 200 times measures one document, not 200 attempts

The second reason large series say so little is mundane. Anyone who manages 200 applications is almost always working from one master dossier with the salutation swapped. The series then measures a single text repeated 200 times rather than 200 independent attempts. A weakness in it is no longer noise that averages out; it is a constant.

That is what people recruiting into the Swiss market describe with striking consistency: rejections of newcomers are rarely about missing experience, but about how the experience is presented — no line on nationality and permit, language skills without a level, a format from the home market, and no visible link between the career so far and what the advert asks for.

The practical conclusion is unwelcome: lengthening the series changes nothing. One application that answers a specific advert is worth more statistically than ten that answer none, and it is the only kind you can learn anything from afterwards.

Where your 200 stopped — three cuts

Cut the series in three places and it shows you where the cause sits. First cut: no human reaction at all. That points at the targeting, at a stated condition, or at a dossier that is not being read cleanly. Second cut: reactions yes, invitations no. That is fit to the individual advert, and only a dossier that visibly picks up the advert's requirements moves it. Third cut: invitations yes, offer no. The document has left the stage by then and the work is in the interview.

For the cuts to work you need three notes per application, none of which concern your document: who received it — the employer itself, an agency or a public administration —, which language the advert was written in, and at which point the no arrived. Who actually decided in your series, and why a rejection from a staffing agency means something different from one by the employer, is worked out from our stock on the question «Why do I only get rejections?».

An example of how fast the picture turns: 200 applications, of which 60 went to adverts demanding two working languages where you hold one, another 12 to apprenticeships and internships, and most of the remainder to positions outside your field because the field itself had nothing more to offer. What looked like a series of 200 is then a series of perhaps 40 genuinely fitting applications — and 40 without a reply, at a true rate of five per cent, is entirely normal in about twelve out of a hundred cases.

What to do now

Narrow before you widen. Take the last twenty adverts you applied to and strike every one that carried a stated condition you do not meet — language level, permit, qualification, years of experience. What remains is your actual slice. If it holds fewer than ten adverts a month, the next decision is about region or occupation, not about your CV.

Then test the dossier against a single advert rather than «in general». A CV is never good or bad in itself; it is fitting or unfitting to one text that somebody wrote. Our CV check reads the advert and your dossier side by side and shows which required points do not appear in the document at all — the examination that a series of 200 can never substitute for.

And if the volume comes from an agreement with your RAV or ORP: how many job-search efforts are required, and where that number comes from, is answered on «How many job applications does the RAV require per month?» with the federal guidance. Either way the efforts still have to be reported to your RAV and recorded the prescribed way to count — a tool can fill the form for you, it cannot replace that step.

Where these numbers stop

Our slices are constructions: occupational fields come from multilingual keyword lists over job titles, regions from the advert's stated location. A narrower word list moves the monthly figures; the order of magnitude holds. Ticino carries a very small share of the stock, so there the values describe the structure rather than the second decimal place.

The language requirements are levels detected in the advert text and they are not error-free — the raw data occasionally records a programming language as a «language». We therefore report them as shares of the adverts that state a requirement at all, not as shares of the whole market. Our stock is also not the whole Swiss market: what is never advertised appears in none of these numbers.

The probabilities are model calculations. They assume every application is an independent attempt, which stops being true as soon as several of them land at the same employer. For your personal situation the RAV remains the competent authority, and for the question of whether your document is the problem, a check against one concrete advert is the shortest route to an answer.